interview

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.80). The skill's Tools section explicitly instructs the agent to perform external research (e.g., "Use MCPs (context7, exa) for external research" and "Web search for prior art, standards, best practices"), meaning the agent will fetch and read public web or user-generated content as part of its workflow, exposing it to untrusted third-party input.